Swadish lists are apparently drawn from several parts of speech of a given language and select vocabulary for lower and higher registers of fluency in a language [please correct me if this is incorrect].
But if correct, then can someone identify a methodology used (beyond just identifying word order) to create such lists?
The goal is to construct written assessment tools/instruments to be administered orally for different Mayan languages from Meso-America. I do not assume a standardized dialect for Mayan languages, indeed some have marked differences. Such dialect differences could be identified as well, if the method uses samples from known dialect regions, the Swadish List then could be used to also distinguish dialect. Languages of interest: Mam, Q'anjobal, Popti, Chuj, Quiche, Awakateco, Ixil, and Achi.
In context of the language of the Sabah people, I believe that it is possible to construct a swadesh list from oral language. A link is provided for you to have a look.
http://studentsrepo.um.edu.my/5575/
Robert,
Thank you for sharing this deep example of differences in register and dialect. A very good study. I assume Sabah is the oral language you refer to . . .
It appears Jane Kon Ling's word list of 150 words (in appendix Bi) was used to detect phonetic register (along with three other methods). Was this a modified Swadish list? Jane apparently did not use IPA for phonemes. Are English language phonemes familiar enough or used in the Sabah, that IPA was not needed?
So to construct an assessment, what level (percentage) of response is needed to just detect a speaker of language X? Dialect differences between the Swadish list dialect chosen, and the speaker's dialect, but the length of the word list should be able to overcome dialect differences enough to determine if the speaker speaks that language. I am thinking that the 100 word Swadish list may be less laborious than the 207 word list.
We are in the context of 7-10 indigenous Mayan languages where there may be no real common language between the person carrying out the assessment and the speaker, or a very limited Spanish for the indigenous language speaker as a bridge. In that context, any suggestion on how to administer the Swadish list as an assessment? Would a short query be better using Swadish list words in a few sentences?
I can be reached at: [email protected]
The use of IPA for phonemes is less preferred in Malaysia largely because the country subscribes to the Malay language as its national language; alphabets with accented characters are not one of our strength.
With regards to the use of a 100-word list or the 207-word list, I guess it depends on the interviewer. Since traditional ecological knowledge would have an impact on indigenous peoples' language, it would be pragmatic to use the 100-word list. The 207-word list seems more suited when the interviewer has a good concept of the traditional ecological knowledge of his or her interviewee that the interviewer could differentiate the interviewee from the objects of his or her surrounding. After all, talking with indigenous peoples most times seem like story-telling.
Robert,
Thank you for your observations. I have spoken with another linguist who suggests that Swadish lists over time are less accurate than other types of assessment for language fluency/register/knowledge. Though I can see how for ecological information, the biome and climate specific language would be rich indeed if the 207 word list was used (or modified) and such an exercise could help document knowledge where it is relational and specifies functions of the interplay between flora and fauna.
Ror the original purpose of my inquiry I will turn to a different method, using geographic information to identify language variants for the purpose of matching language speakers with interpreters from informed identification of their language. One tidbit I did not know was that Swadish actually studied Mayan Languages in Guatemala/Mexico (Meso-America) which are where Mayan languages reside which are our languages of interest.
Blake:
For background on Morriss Swadesh and his work in Mesoamerica, I recommend taking a look at volume 5 of the Handbook of Middle American Indians, on linguistics, if you should find yourself in a decent academic library with a few spare minutes at hand. He has an chapter in this volume where he describes and summarizes his lexicostatistical work, using the 100 word list to compare related languages and assess their degree of relatedness in glottochronological terms.
Davis,
Since the post, a linguist explained to me that Swadish had worked in Mayan languages; I assumed it was in Indo-European languages. Silly me. And since then, I am fairly convinced that there is a more direct route for creating an oral language assessment for the purposes I have in mind. Thank you for the tip on his work, I'll take a look at it. I have had my academic credential as an "independent contractor" researcher at the College of Law, renewed. Library privileges are again mine. The Swadish lists though may come in handy for a second purpose.
P.S. Swadesh was bullied out of U.S. academia during the McCarthy era, so he moved to Mexico City in 1956 and spent a decade there teaching and researching native languages. This was great for Mexican anthropology, since he founded a vigorous tradition of linguistic research that continues to the present.
Wisconsin gives (Progressive era, Univ. of Wisconsin) and it takes away (McCarthy). Not much has changed.
Swadish apparently turned lemons into limonada, and I love limonada..